mineral detection and mapping using band ratioing and

International Journal of Scientific & Engineering Research, Volume 5, Issue 12, December-2014 1100 ISSN 2229-5518

IJSER © 2014 http://www.ijser.org

Mineral Detection and Mapping Using Band Ratioing and Crosta Technique in Bwari Area

Council, Abuja Nigeria. Sadiya T. B., Ibrahim O., Asma T. F., Mamfe V., Nsofor C.J., Oyewmi A. S., Shar J.T., Sanusi M., Ozigis M.S.

Abstract- Landsat 7ETM+ of Bwari local government area of Abuja federal Capital Territory located in the middle belt of Nigeria was used to detect and map locations of hydrothermal alterations. Image processing methods used includes image rectification, spatial and spectral enhancement, band ratio, false colour composition (FCC) and Crosta technique. Band ratios 3/1 and (4/5 – 4/3) suggested the presence of ferric iron minerals and hydroxyl minerals respectively. Clay mineralization was detected using band ratio 5/7. While false colour composite of bands 7:4:2 was employed to delineate potential locations of hydrothermal alterations. In both the band ratio and false colour composite there was interference from vegetation which affected the results. Subsequently, Principal Component Analysis (PCA) and Crosta technique were employed to supress the interference. Resultant grey tone images from PCA shows white pixels depicting iron-oxide and hydroxyl mineral deposits. To enhance the location of mineral deposits, false colour composite of the resultant images and a sum of the two images were displayed as RGB respectively. Hydrothermal alteration zones hosting the mineral deposits were identified. The identification of the hydrothermal alteration zones were correlated with the geological map and the ground-truth coordinates of already existing mines obtained from the study area.

Keywords- Band ratio, Colour Composite, Crosta technique, hydrothermal alteration zones, Mineral deposit, Principal Component Analysis, Remote sensing, and Landsat 7 ETM+.

—————————— ——————————

1 Introduction Nigeria is a country blessed with enormous mineral

resources deposited all over the country [1]. Unfortunately,

these mineral resources contribute only about 1% to the

country’s Gross Domestic Product (GDP) [2][3] due to

government negligence of the mining sector and the

proliferation of small scale illegal mining operations.

The small scale mining operations are carried out by

untrained miners and the operations themselves are

unregulated and without guidelines. This usually results in

environmental degradation and loss of minerals [4].

Therefore, there is the need for a mineral resource

inventory to aid the government in implementing better

laws, regulations and mitigation to guide small scale

miners in carrying out safer mining activities and also

increase the country’s GDP.

Considering that different types of mineral deposits are

found all over the country, traditional method of mineral

mapping can be costly and time consuming. A more

modern method involves the use of remote sensing for the

detection, identification and mapping of hydrothermally

altered rocks which are a pointer to the locations of mineral

deposits [5]. Remote sensing is the acquisition, processing

and interpretation of images and related data which are

acquired from airborne objects and satellites that record the

interaction between object/phenomena and electromagnetic

energy [6]. Remote sensing can detect and map

hydrothermally altered mineral which lie on the earth’s

surface. Large areas of interests are mapped at a relatively

fast pace while at the same time keeping the financial costs

down [7]. Hydrothermal mineralization alteration is a

process which alters the mineralogy and chemistry of the

host rocks resulting in producing mineral assemblages

IJSER

http://www.ijser.org/



which vary according to the location, degree and duration

of the alteration processes [7]. These minerals are unique in

the way they absorb or reflect light along the

electromagnetic spectrum due to their different chemical

compositions, thus making it possible for minerals to be

detected and mapped using remote sensing.

Hydrothermally altered minerals (e.g kaolinite, allunite,

muscovite etc.) are in most cases indications of a deposit of

precious or economically viable mineral [8].

Several authors have analyzed and interpreted the remotely

sensed hydrothermally altered mineral deposit data

through band ratioing. For example Abdelsalam et al., in

2000 [9] used ETM+ band ratios (5/7, 4/5 and 3/1) in RGB to

map the Beddaho alteration zone in northern Eritrea. [10]

used Landsat ETM+ data for mapping gossans and iron rich

zones exposed at Bahrah Area, Western Arabian Shield,

Saudi Arabia. He found out that 4/5 ratio discriminate well

between the gossans and iron-rich zones. Ratio 4/5 was also

used to differentiate between the gossans, iron-rich rocks

and surrounding rocks. The use of distinct spectral

reflectance and absorption feature in ETM+ band ratio 5/7

and 3/l highlights the presence of clay and iron minerals

which is used to detect and map hydrothermal alteration

zones by [6].

In Nigeria, there has been very little research in using

remote sensing techniques to identify mineral deposits.

One of the few carried out was [11] who applied band ratio

technique and colour composite to map location of

hydrothermal alteration within the Maru schist belt of

Northwest Nigeria. In view of this, the aim of this paper is

to detect and map the distribution of hydrothermally

altered mineral deposit zones in the peri-urban area of

Abuja as a means of initiating a mineral deposit inventory

database for Nigeria.

1.1 Geological Setting of Abuja

Nigeria lies within the zone of Pan-African reactivation

(ca.550 Ma) to the east of the West African Craton, which

has been stable since approximately 1600Ma. This mobile

belt extends from Algeria across the Southern Sahara into

Nigeria, Benin and Cameroon. Rocks of the Nigerian

Basement Complex which is part of the Pan African Mobile

Belt are intruded by Mesozoic ring complexes of Jos area

and overlain unconformably by Cretaceous to Quaternary

sediments forming the sedimentary basins. Three broad

lithological groups have been distinguished in the Nigeria

Basement Complex: A polymetamorphic, Migmatite-Gneiss

Complex with ages ranging from Liberian (ca. 2800 Ma) to

Pan-African (ca. 600Ma). The metavolcanosedimentary belt

which is essentially made of the schists of low grade

metasediments of greenschist facies of Proterozoic age (iii)

the syntectonic to late tectonic granitoids of Precambrian to

Lower Palaeozoic age which intrude both the gneiss

complex and the schist belt [12][13][14][15][16].

The study area Abuja is predominantly underlain by the

pre-Cambrian basement complex rocks. McCurry in 1976

[17] defined the basement complex of Nigeria as the

reactivated ancient crystalline rocks which formed a suit of

migmatite, gneiss and granite grouped as a single

petrogenetic unit. Rahaman in 1988 [13] pointed out that the

metamorphism of the basement rocks of Nigeria was

polycyclic. Rahaman in 1976 [18] argued that about 60% of

the basement complex rocks in north central Nigeria are

made up of migmatite and gneissic rocks which have

undergone at least three stages of deformational trends

(metamorphism) before their present state. The local

lithological units in the study area are migmatite-gneiss,

granite, and schistose gneiss. Granite is the most wide

spread rock unit occurring in several locations; they are

IJSER




porphyritic and of medium-coarse-grained texture,

Granites mostly occur as intrusive, low-lying outcrops

around the gneiss. They are severely jointed and fairly

incised by quartz veins with fractures trending northerly in

the NNE-SSW, NNW-SSE directions.

1.2 Study Area and Geological Setting

2.0 Methodology

Throughout this study, two ground-truths exercises were

carried out around the locations of past and present mining

sites on 4th-July, 2014. It was found out that most of the

mining sites are exploring granite for construction in

economical quantities while other sites are precious

minerals (e.g. tantalite, tourmaline, marble, quartz, etc.). Fig

6d shows image of a location where mining activity is

currently going on.

2.1 Data Used and Pre-processing

Landsat 7 Enhanced Thematic Mapper Plus (ETM+) data

scenes number 189/54(Path/Row) obtained on 1st-Jan-1998

when the weather was relatively dry was used. Geologic

map of the study area was obtained from Nigerian

Geological Survey [19] this will be use to compare locations

of hydrothermal alteration zones detected with using

remote sensing technique with locations of mineralization

given in the geological map. The image was corrected for

geometric distortions and projected to geographic

(lat./long.) coordinate system and WGS-84 datum. The

study area was subsetted from raster image (ETM+ data)

using the shapefile of the Bwari LGA. The image was

atmospherically corrected using Internal Average Relative

Reflectance (IARR) method [20] and converted to surface

reflectance.

2.2.0 Techniques Used 2.2.1 Band Ratio

Considering that each object has a unique spectral

reflectance curve in each wavelength of the electromagnetic

spectrum, band ratio can be used to emphasize the anomaly

of target object by determining the band at which

reflectance is high or point of highest absorption [21][22].

Band ratio also reduces the effect of topography, hence,

enhancing the differences between spectral responses of

each band [23]. Landsat ETM+ bands 3/1 can detect the

IJSER




smallest amount of Ferric iron-bearing surfaces of

hydrothermally altered rocks, sedimentary rocks,

metamorphic rocks containing weathered, iron-bearing

mafic minerals such as hornblende, biotite, sand deposits,

and alluvium derived from such rocks will be identified

with this index [24]. Clay mineralization is detected using

the ratio 5/7 indicating that band 5 has reflectance in

contrast to band 7 which has a high absorption. While

subtracting band ratio 4/3 from 5/7 is done to remove the

effect of vegetation interference in the detection of clay

mineralization. Band ratio 5/4 is used in order to detect

locations with hydroxyl mineralization, which is another

indicator of hydrothermally altered zones [24][9].

2.2.2 Principal Component Analysis (PCA) and Crosta Technique

PCA is usually used for detecting and mapping the

distribution of alteration in metallogenic provinces [21].

PCA is a multivariate statistical method that analyzes the

Eigenvector loadings (eigenvalues) of a multispectral

satellite image to determine the unique spectral response of

different minerals and rocks contained in the image called

Crosta Technique [5][7].

Sabins in 1996 [6] successfully adopted a methodology

where PCA of Band sets 1347 and 1547 were run separately

to determine the presence of iron-oxide and hydroxyl-rich

minerals respectively. PC1 mapped albedo and

topographic information while PC2 displayed the VIS/IR

versus SWIR bands in contrasting signs [25].

The eigenvectors of PCs 3 & 4 were checked to see if they

contain substantial loadings in opposite signs from input

bands of 1 & 3 and 5 & 7, since these band pairs are

expected to display contrasting response for iron and

hydroxyl-rich minerals respectively [26].

3.0 Result and Analysis

Fig 2: Bands 4:7:2 false colour composite of Landsat ETM+ subset in RGB showing study area in Bwari Area Council. Using false colour composite, hydrothermal zones whose

mineral spectral signatures are in near infrared (IR) to mid-

infrared can be detected using ETM+ [27]. In Fig 2 altered

minerals are seen in pale green, white and pale purple from

creating a false colour composite of bands 4:7:2 in RGB

respectively.

3.2.0 Band Ratios

Band ratio 3/1 highlighted areas in which any ferric iron

mineral occurs pervasively or as coatings (Fig 3a). In Fig 3b

band ratio (4/5 – 4/3) depicted hydroxyl mineral due to

hydroxyl’s high reflectance in band 4 and absorption

minima in band 5. Subtraction of ratio 4/3 was carried out

to reduce the effect of vegetation interference in the mineral

detection. Clay mineralization was detected using the ratio

5/7. This shows that band 5 has reflectance in contrast to

band 7 which has higher absorption property (see Fig 3c).

IJSER




While ratio (5/7 – 4/3) was used to subtract the effect of

vegetation interference in the detection.

Due to spectral properties of the minerals, the colour

composite of (R-(3/1), G-((5/7)-(4/3)) and B-((4/5)-(4/3))) was

created to delineate the locations of hydrothermally altered

rocks in pinkish to red and blue colours while vegetation

appears green (Fig 4). To remove the effect of vegetation

from interfering with mineral detection and mapping,

statistical techniques called Principal Component Analysis

(PCA) and Costa Technique were employed.

Fig 3a: Band ratio (3/1) Showing of Iron-Oxide mineralization in white

Fig 3b: Band ratio ((5/7)-(4/3))

showing few pixels of clay Mineralization in white.

Fig 3c: Band Ratio ((4/5)-(4/3)) showing Hydroxyl mineralization in white pixels.

Fig 4: colour composite of (R-(3/1), G-((5/7)-(4/3)) and B-((4/5)-(4/3))) hydrothermally altered rocks in pinkish to red and blue colours while vegetation is in green.

3.3.0 Principal Component Analysis and Crosta Technique

IJSER




The Principal Component Analysis was performed on two

sets of band combination; 1457 and 1347. The PCA

eigenvector values are shown in Tables 1 & 2. In Table 1,

PC4 appears to have mapped iron-oxide minerals in dark

pixels as observed in Band 3 which has high eigenvector

value of -0.901413. So also in Table 2, PC4 appears to have

mapped hydroxyl minerals in dark pixels. This is observed

in Band 5 having a higher eigenvector value (-0.580019)

compared to PC3 eigenvector value (-0.337910) in the same

Band.

To display the iron-oxide and hydroxyl minerals in bright

pixels, the negative of each image was taken (Figs 5a & 5b).

In the resulting image bright pigmentation (Fig 5a) was due

to the high reflectance of iron-oxide in band 3 and

absorption minima in band 1; this image is called F-image.

The bright pixels of hydroxyl mineral (Fig 5b) was due to

the high reflectance in band 5 and absorption in band 7

[24][25]; this image is called H-image.

Eigenvector Band 1 Band 3 Band 4 Band 7 PC 1 0.445965 0.381777 0.712609 0.384122 PC 2 0.177587 -0.142172 0.434724 -0.871358 PC 3 -0.826092 -0.146589 0.530668 0.120308 PC 4 0.295211 -0.901413 0.146950 0.280555 Table1: Principal component analysis of bands 1347 for Fe-oxides mineral Eigenvalues Band 1 Band 4 Band 5 Band 7 PC 1 0.382319 0.610790 0.607731 0.333812 PC 2 0.297129 0.579880 -0.424331 -0.628806 PC 3 0.823226 -0.343280 -0.337910 0.300457 PC 4 -0.296388 0.415733 -0.580019 0.634742

Table2: Principal component analysis of bands 1457 for hydroxyl minerals

False colour composite of the images corresponding to F-

image, H-image and the sum of both (i.e. F-image + H-

image) are displayed as RGB composite (i.e. F-image (R), H-

image (G), F+H (B)) to allow the identification of

hydrothermally altered rocks (Fig. 5) in pinkish to red

colour.

PCA of Bands 1347 PCA of Bands 1457

(5a). PC 4 : F-Image

(5b) PC 4: H-Image

Fig 5a & 5b: Showing Iron-oxide and hydroxyl minerals in

white pixels.

Fig 6: Colour composite of R-(F-image), G-(H-image) and the sum of B-(F+H). (b) Shows pixels of ground-truth site 1&2. (C) Represent pixels of ground-truth sites 3&4. (d) Image of one of Katampe quarry sites.

IJSER




Fig 7: comparing remote sensing data with existing geological map on location of mineral deposits. Red polygons indicate locations corresponding to known locations on geological map while black polygons represent newly detected sites.

4.0 Discussion

False colour composite, band ratios and principal

component analysis were utilized in order to spectrally

enhance the response of different minerals such as iron-

oxide, clay and hydroxyl which represent hydrothermal

alteration zones such as the ones in Katampe (Fig 6d). From

principal component analysis; in both tables PC1 have

positive eigenvalue loadings which makes material

identification using spectral signature impossible [27].

However, PC4 (Fig 5a) contains the spectral information

depicting the iron-oxide bearing minerals in the study area

which is much more than the amount detected using band

ratio of 3/1. This is because PCA excludes the effect of

vegetation signature interference [28]. It can be seen that

the iron-oxide concentrations formed little clusters,

occurring mostly in the northwest, north-central and

eastern portions of the image. These areas coincide with

locations being mined for granite, tantalite, marble and

other precious stones (as observed during the ground

verification exercise) and also with the geological map (red

polygons in Fig 7). So also from Fig 7 the black polygons

represent other detected potentially interesting sites that

are absent in the geological map. These sites are subject to

further investigation.

Both the PC4 (Fig 5b) and band ratio (Fig 3c) indicate a

large concentration of hydroxyl bearing minerals probably

due to sericitisation during the alteration process, with the

highest concentration in the north-northwest and north-

east. This explains why there are few pixels of clay

mineralization; this is because concentration of clay

mineralization varies with the concentration of hydroxyl

mineral [29]. Even though the band ratio colour composite

(i.e. RGB) highlighted the spatial distribution of

mineralized surfaces (Fig 4), it still exaggerated the extent

of mineralization by depicting un-mineralized rock-outcrop

and/or bare surface as hydrothermal alteration zones. This

might be due to the effect of vegetation interference and

other lithologies having same spectral signatures with the

actual mineral deposits.

Fig 7 reveals corresponding mineral deposits location on

satellite image and those observed on the geological map.

Hydrothermal deposits are products of hydrothermal

processes. These deposits form economic reserves when

they are concentrated in veins, and other voids. Veins and

lodes consist of the infillings of fissures and fractures

developed in the outer part of an intrusive body or in the

surrounding roof and wall-rocks. Veins which contain

metalliferous minerals are termed lodes [30]. Minerals such

as quartz, pyrite, topaz, and tourmaline are commonly

associated with metalliferous minerals in lodes. Just as

obtained from the geology of the study area, metals which

IJSER




are commonly associated in this way with acid rocks

include copper, lead, zinc, arsenic, tungsten, gold and

silver. In the study area, tin exists in the form of casseterite

(SnO2), tantalite [(Fe,Mn)Ta2O6)], the primary source of

tantalum is presently being mined. Quartz in granites and

gravels are majorly being mined for construction purposes

while marble and tourmaline are being mined in small

quantities by artisans.

5.0 Conclusion

Application of Crosta’s Technique with two sets of four

spectral bands (i.e. tables 1 & 2) has proven to be more

efficient for the mapping of hydrothermal alteration zones

in the form of iron-oxide and hydroxyl minerals. This can

be seen in Fig.6 where locations of mineralization are

represented in pinkish to red colouration.

This study has established that the combined use of spatial

and spectral resolution of satellite data could result in the

detection and mapping of locations with known and

unknown mineral deposit. Intrusion-related mineralization

indicators such as iron-oxide, clay and hydroxyl minerals in

the Bwari local government area mining locations have

been detected using Landsat7 ETM+ dataset and the

technique used in this investigation. The techniques used

allowed the removal of spectral effect of vegetation from

the EMT+ images resulting in the detection of the mineral

deposit. This research could be made better by using

hyperspectral images to determine the exact indicator

mineral deposit in the study area.

References

[1] O. Uzoka, “Chairman, House Committee on Solid Minerals, National Assembly, Nigeria”, Interview with Thisday Newspapers, available from

: http://www.thisdayonline.com/archive/2001/08/26/200110826news12.html, (March 17, 2002).

[2] L. Awute “Permanent Secretary, Ministry of mines and steel Development, Nigeria”. Interview with Vanguard newspaper at the Nigerian-Brazil Investment forum in Abuja. Vanguard Newspaper, (2013). Available from: http://www.vanguardngr.com/2013/02/nigeria-loses-n8trn-to-illegal-gold-mining/. (june 29, 2014)

[3] Nigeria: Mining – Overview, 2014. Available from: http://www.mbendi.co.za/indy/ming/af/ng/p0005.htm, (20 April, 2014)

[4] I. Aigbedion, & S.E. Iyayi, “Environmental effect of mineral exploitation in Nigeria”. International Journal of Physical Sciences, international Journal of Physical Sciences, 2 (2), 033-038, 2007.

[5] A.P. Croster, C.R.S. Filho, F. Azevedo & C. Brodie, “Targeting key alteration minerals in epithermal deposits in Patagonia, Argentina, using ASTER imagery and principal component analysis”. INT. J. Remote Sensing, 24(21), 4233–4240, 2003.

[6] Sabins, F.F. “Principles and Interpretation” in W.H. Freeman (3rd ed.), Remote Sensing (New York NY: 494), 1997.

[7] M.B. Mia & Y. Fujimitsu “Mapping hydrothermal altered mineral deposits using Landsat 7 ETM+ image in and around Kuju volcano, Kyushu, Japan”, Journal of Earth System Science, 121(4), 1049–1057, 2012.

[8] J. B. Torre, J. G. R. Ayuga, J. Bonatti, M. M. Sacristán & R.M. Marín “Detection of hydrothermal alteration using a principal component analysis applied to hyperespectral hymap data on the Turrialba Volcano, Costa Rica”, International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, XXXIX-B7, 2012.

[9] M.G. Abdelsalam, R.J. Stern, & W.G. Berhane ”Mapping gossans in arid regions with Landsat TM and SIR-C images: the Beddaho Alteration Zone in northern Eritrea”, International Journal of African Earth Science, 30(4), 903-916, 2000.

[10] A. Madani “Utilization of landsat etm + data for mapping gossans and iron rich zones exposed at Bahrah Area , Western Arabian Shield , Saudi Arabia”, 35-49, 2009.

[11] E. A. Kudamnya, W.T. Andongwa, & J.O. Osumeje, “Hydrothermal Mapping of Maru Schist Belt, North-Western Nigeria Using Remote sensing Technique”. International journal of civil engineering, 3(1), 59-69, 2014.

[12] K.C. Burke & F.J. Dewey. Orogeny in Africa. In: Dessauvagie, T.F.J. and Whiteman, A.J. (Ed.), African Geology, (Ibadan University Press, Ibadan) 583-608, 1972.

[13] M.A. Rahaman, “Recent advances in the study of basement complex of Nigeria”. In Precambrian Geology of Nigeria, Nigeria Geological Survey, , 11–43, 1988.

[14] E. Annor, & S.Freeth,” Thermo-tectonic evolution of the basement complex around Okene, Nigeria, with special reference to deformation mechanism, Precambrian Research” , 28:269-281, 1985.

[15] I. Odeyemi,”Lithostratigraphy and structural relationships of the upper Precambrian metasediments in Igarra area, southwestren Nigeria, Precambrian Geology of Nigeria”, Nigeria Geological Survey, 111–125, 1988.

IJSER


http://www.thisdayonline.com/archive/2001/08/26/200110826news12.html

http://www.thisdayonline.com/archive/2001/08/26/200110826news12.html

http://www.vanguardngr.com/2013/02/nigeria-loses-n8trn-to-illegal-gold-mining/

http://www.vanguardngr.com/2013/02/nigeria-loses-n8trn-to-illegal-gold-mining/

http://www.mbendi.co.za/indy/ming/af/ng/p0005.htm



[16] E. Annor, ”A note on the geology of the Isanlu area, in the Egbe-Isanlu schist belt, S.W. Nigeria”, Journal of Mining and Geology, 47–52, 1995.

[17] P. McCurry,”The Geology of the Precambrian to lower Paleozoic rocks of northern Nigeria”, In K. C. (ed.), A review In:Geology of Nigeria, (Lagos, Nigeria: Elizabethan Publishing Co), 15-29, 1976.

[18] C. Ajibade,” Proterozoic crustal development in the Pan african regime of Nigeria” C. A. (Ed.), Geology of Nigeria. Lagos, (Elizabethan Publishing Co), 57–69, 1989.

[19] Nigeria Geological Survery, 2014. Available:http://wrf.nigeriagovernance.org/organizations/view/870, (March 16, 2014)

[20] F.A. Kruse, G.L. Raines, & K. Watson. Analytical techniques for extracting geologic information from multichannel airborne spectroradiometer and airborne imaging spectrometer data. In Int. Symp. on Remote Sensing of Environment 4th Thematic Conference on Remote Sensing for Exploration Geology, Ann Arbor, Michigan, ERIM, 1-4 April, 309–324, 1985.

[21] R. Nouri, M.R. Jafari, M. Arain, & F. Feizi, Hydrothermal Alteration Zones Identification Based on Remote Sensing Data in the Mahin Area, West of Qazvin Province, Iran, World Academy of Science, Engineering and Technology, V6, 2012.

[22] T. San, E. O. Sumer, & B. Gurcay,”Comparison of band ratioing and spectral indices methods for detecting alunite and kailinite minerals using ASTER data in Biga region, Turkey, Geo-Imagery Bridging Continents” XXth ISPRS Congress, Istanbul, Turkey, July 2004.

[23] A. Ghulam, R. Amer, & T. M. Kusky,”Mineral Exploration and Alteration Zone Mapping in Eastern Desert of Egypt using Aster Data”, ASPRS Annual Conference San Diego, California, 2010.

[24] B.W. Rockwell, “Automated mapping of mineral groups and green vegetation from Landsat Thematic Mapper imagery with an example from the San Juan Mountains Colorado: U.S. Geological Survey Scientific” Investigations http://pubs.usgs.gov/sim/3252/, 2010.

[25] N. A. Kenea, & H. Haenisch, “Principal component analysis for lithologic mapping and alteration mapping: Examples from the Red Sea Hills”. International achieves of photogrammetry and remote sensing. XXXI, part B7, 1996.

[26] W. P. Loughlin,”Principal component analyses for alteration mapping”, Photogrammetric Engineering and Remote Sensing, 57 (9), 163-1169, 1991.

[27] J.R. Jensen,”Introductory Digital Image Processing”. (Pearson Prentice Hall: Upper Saddle River), 2005.

[28] A.P., Crosta, & J.M.M. Moore,”Enhancement of Landsat Thematic Mapper imagery for residual soil mapping in SW Minais Gerais State, Brazil: a prospecting case history in Greenstone belt terrain”, Proceedings of the 7th ERIM Thematic Conf.: Remote Sensing for Exploration Geology, 1173–1187, 1989.

[29] B., Pour, M., Hashim, & J. B. Genderen,”Detection of hydrothermal alteration zones in a tropical region using satellite remote sensing data: Bau goldfield, Sarawak, Malaysia”, Ore Geology Reviews 54:181–196, 2013.

[30] Blyth, F.G.H., & De Freitas, M.H. (1984). Geology for Engineers (7th ed), (Edward Arnold (Publishers) Ltd, London).

IJSER


http://wrf.nigeriagovernance.org/organizations/view/870

http://wrf.nigeriagovernance.org/organizations/view/870

http://pubs.usgs.gov/sim/3252/

mineral detection and mapping using band ratioing and

Documents